Data Science Glossary for Beginners

Data Science · Beginner Glossary

Data Science Glossary for Beginners

AK Ankit Kaushik · Lead Trainer, Data Engineering & AI | Updated 28 August 2026 | 14 min read Free · Beginner Guide

A beginner's path through data science terms — from data basics to machine learning and statistics.

Quick summary — data science glossary for beginners

Data science has a language of its own. This glossary breaks down the most important terms into plain English so you can start learning without feeling lost.

In this guide you will learn:

Data basics — dataset, feature, label, target.
Machine learning terms — model, training, inference, overfitting.
Statistics terms — mean, variance, p-value, correlation.
Data engineering terms — pipeline, ETL, data warehouse.
Evaluation terms — accuracy, precision, recall, F1.

SECTION 01Data basics

These are the foundational terms you'll see in almost every data science project:

Term	Definition	Example
Dataset	A collection of data organised for analysis	A CSV of customer purchases
Feature	An input variable used to make predictions	Age, income, location
Label / Target	The output variable you want to predict	Whether a customer churns
Row / Record	A single observation in a dataset	One customer's data
Data cleaning	Fixing or removing incorrect or missing data	Handling null values

Key insight: Most data science work starts with understanding and cleaning data — the terms above are your first building blocks.

SECTION 02Machine learning terms

Machine learning has its own vocabulary. Here are the essentials:

Term	Definition	Why it matters
Model	A mathematical representation learned from data	Used to make predictions
Training	The process of teaching a model using data	Improves model accuracy
Inference	Using a trained model to make predictions	Real-world predictions
Overfitting	When a model memorises training data too closely	Poor performance on new data
Supervised learning	Learning with labelled data	Classification and regression
Unsupervised learning	Finding patterns in unlabelled data	Clustering and dimensionality reduction

Core machine learning terms:

Model        → a learned function that maps inputs to outputs
Training     → fitting a model to data
Inference    → using a trained model to predict
Overfitting  → model learns noise, not signal
Underfitting → model is too simple to capture patterns

Tip: Always validate on data the model hasn't seen.

Types of machine learning:

Supervised   → labelled data (classification, regression)
Unsupervised → unlabelled data (clustering)
Semi-supervised → a mix of labelled and unlabelled
Reinforcement → learning through rewards and penalties

Start with supervised learning — it's the most common
in entry-level data science roles.

ml-glossary.md

Key insight: Understanding the difference between training and inference is critical — it's a common interview question.

SECTION 03Statistics terms

Statistics is the backbone of data science. Here are the terms you must know:

Term	Definition	Practical use
Mean	The average of a set of numbers	Measuring central tendency
Median	The middle value when data is sorted	Better for skewed data
Variance	How spread out numbers are from the mean	Measuring variability
Standard deviation	The square root of variance	Interpretable spread measure
Correlation	How two variables move together	Feature selection
p-value	Probability of observing results by chance	Hypothesis testing

Pro tip: You don’t need a statistics degree to start — but understanding mean, variance, and correlation will make your analysis far more reliable.

SECTION 04Data engineering terms

Data science depends on clean, well-engineered data. These terms help you understand the pipeline:

Term	Definition	Why it matters
Data pipeline	A series of steps to move and transform data	Automates data flow
ETL / ELT	Extract, Transform, Load data between systems	Prepares data for analysis
Data warehouse	A central repository for structured data	Enables fast analytics
Data lake	Storage for raw, unstructured data	Flexible, scalable storage
Schema	The structure of a dataset (columns, types)	Ensures consistency

Key insight: Data scientists spend a lot of time working with data engineers — knowing these terms helps you collaborate effectively.

SECTION 05Model evaluation terms

Once you build a model, you need to evaluate it. These terms are essential:

Term	Definition	When to use
Accuracy	Percentage of correct predictions	Balanced classes
Precision	Of predicted positives, how many were correct	When false positives are costly
Recall	Of actual positives, how many were found	When false negatives are costly
F1 score	Harmonic mean of precision and recall	Imbalanced datasets
Confusion matrix	A table of predicted vs actual values	Diagnosing model errors

Pro tip: Accuracy can be misleading when classes are imbalanced — use precision, recall, and F1 for a clearer picture.

SECTION 06Interview Q&A — data science glossary

Q1What's the difference between a feature and a label?

A feature is an input variable used for prediction, while a label (or target) is the output variable you're trying to predict.

Q2What does overfitting mean?

Overfitting happens when a model learns the training data too closely, including its noise, and performs poorly on new, unseen data.

Q3When should you use median instead of mean?

Use the median when your data has outliers or is skewed — it's a more robust measure of central tendency in those cases.

Q4What's the difference between precision and recall?

Precision measures how many predicted positives were actually correct, while recall measures how many actual positives were found by the model.

Q5What is an ETL pipeline?

ETL stands for Extract, Transform, Load — a process for moving data from sources into a data warehouse after cleaning and reshaping it.

SECTION 07Test yourself — data science glossary quiz

Five questions. No sign-up.

0 / 5

Pick an answer to see why it is right or wrong.

SECTION 08Frequently asked questions

Do I need to memorise all these terms?

Not immediately — start with the data basics and machine learning terms, then expand as you learn. Repetition and practice will make them stick.

Which terms are most important for interviews?

Overfitting, precision vs recall, and the difference between supervised and unsupervised learning are frequently asked.

How can I learn data science as a beginner?

Start with Python, statistics, and SQL, then build small projects. This glossary will help you follow tutorials and documentation.

What's the difference between data science and data engineering?

Data engineers build and maintain data pipelines and infrastructure, while data scientists analyse data and build models.

How do I remember all these terms?

Use them in practice — write code, build projects, and explain concepts to others. Regular use is the best way to retain them.

What I would do in the first 30 daysLayoff · 14 min Why sending 500 resumes won't workJob Search · 14 min Rebuild your resume for the AI job marketResume · 15 min Which skills to learn before applyingSkills · 14 min 90-day career recovery roadmapRecovery · 16 min How AI can help you prepare for interviewsAI · 11 min

Classroom & online · Noida

Start your data science journey

Our Data Science Training Course is designed for beginners — learn the terms, tools, and techniques to build your first data science projects.

₹15,500 · full programme ~~₹24,000~~

Explore the course Download the syllabus

Beginner-friendly curriculum
Hands-on projects
Expert mentorship
Weekday & weekend batches

Data Science Glossary for Beginners